Advances in Air Quality Monitoring: A Comprehensive Review of Algorithms for Imaging and Sensing Technologies

Air pollution is a major global concern, leading to serious health problems and environmental damage. This article provides a comprehensive review of historical and current methods used to monitor and predict air quality. It emphasizes the ongoing need for better monitoring techniques. 47 studies are critically analyzed, and computational advancements in air quality monitoring are categorized into sensor-based and image-based techniques. This review reveals that sensor-based algorithmic methods, representing 62% of the reviewed literature, are reliable but often lack flexibility and real-time monitoring capabilities. On the other hand, image-based techniques, while innovative, are limited by the size and diversity of datasets, primarily functioning only during daylight hours. To address these limitations, a hybrid approach that integrates both sensor and image-based methods is proposed. This aims to enhance monitoring by allowing users to visualize pollution levels through an augmented reality layer. The proposed model seeks to provide mobile users with the ability to accurately monitor surrounding air quality by establishing a comprehensive image-based dataset that includes various features not previously considered in existing datasets. This paper examines current methods for monitoring and predicting air quality, focusing on sensor-based and image-based techniques. It analyzes 47 studies to identify strengths and weaknesses of each approach. A hybrid model combining both methods is proposed, aiming to enhance air quality monitoring through augmented reality visualization. The paper concludes by outlining strategies for improving the model's accuracy and usability.image